Before the introduction of antibiotics, patients suffering from acute microbial infections (e.g. tuberculosis or pneumonia) had a low chance of survival. For example, mortality from tuberculosis was around 50%. Although the introduction of antimicrobial agents in the 1940s and 1950s rapidly changed this picture, bacteria have responded by progressively gaining resistance to commonly used antibiotics. Now, every country in the world has antibiotic-resistant bacteria. Indeed, more than 70% of bacteria that give rise to hospital acquired infections in the USA resist at least one of the main antimicrobial agents that are typically used to fight infection (Nature Reviews, Drug Discovery 1, 895-910 (2002)).
One way of tackling the growing problem of resistant bacteria is the development of new classes of antimicrobial agents. However, until the introduction of linezolid in 2000, there had been no new class of antibiotic marketed for over 37 years. Moreover, even the development of new classes of antibiotic provides only a temporary solution, and indeed there are already reports of resistance of certain bacteria to linezolid (Lancet 357, 1179 (2001) and Lancet 358, 207-208 (2001)).
In order to develop more long-term solutions to the problem of bacterial resistance, it is clear that alternative approaches are required. One such alternative approach is to minimise, as much as is possible, the opportunities that bacteria are given for developing resistance to important antibiotics. Thus, strategies that can be adopted include limiting the use of antibiotics for the treatment of non-acute infections, as well as controlling which antibiotics are fed to animals in order to promote growth.
However, in order to tackle the problem more effectively, it is necessary to gain an understanding of the actual mechanisms by which bacteria generate resistance to antibiotic agents. To do this requires first a consideration of how current antibiotic agents work to kill bacteria.
Antimicrobial agents target essential components of bacterial metabolism. For example, the β-lactams (e.g. penicillins and cephalosporins) inhibit cell wall synthesis, whereas other agents inhibit a diverse range of targets, such as DNA gyrase (quinolones) and protein synthesis (e.g. macrolides, aminoglycosides, tetracyclines and oxazolidinones). The range of organisms against which the antimicrobial agents are effective varies, depending upon which organisms are heavily reliant upon the metabolic step(s) that is/are inhibited. Further, the effect upon bacteria can vary from a mere inhibition of growth (i.e. a bacteriostatic effect, as seen with agents such as the tetracyclines) to full killing (i.e. a bactericidal effect, as seen, e.g. with penicillin).
Bacteria have been growing on Earth for more than 3 billion years and, in that time, have needed to respond to vast numbers of environmental stresses. It is therefore perhaps not surprising that bacteria have developed a seemingly inexhaustible variety of mechanisms by which they can respond to the metabolic stresses imposed upon them by antibiotic agents. Indeed, mechanisms by which the bacteria can generate resistance include strategies as diverse as inactivation of the drug, modification of the site of action, modification of the permeability of the cell wall, overproduction of the target enzyme and bypass of the inhibited steps. Nevertheless, the rate of resistance emerges to a particular agent has been observed to vary widely, depending upon factors such as the agent's mechanism of action, whether the agent's mode of killing is time- or concentration-dependent, the potency against the population of bacteria and the magnitude and duration of the available serum concentration.
It has been proposed (Science 264, 388-393 (1994)) that agents that target single enzymes (e.g. rifampicin) are the most prone to the development of resistance. Further, the longer that suboptimal levels of antimicrobial agent are in contact with the bacteria, the more likely the emergence of resistance.
Moreover, it is now known that many microbial infections include sub-populations of bacteria that are phenotypically resistant to antimicrobials (J. Antimicrob. Chemother. 4, 395-404 (1988); J. Med. Microbial. 38, 197-202 (1993); J. Bacteria 182, 1794-1801 (2000); ibid. 182, 6358-6365 (2000); ibid. 183, 6746-6751 (2001); FEMS Microbiol, Lett. 202, 59-65 (2001); and Trends in Microbiology 13, 34-40 (2005)). There appear to be several types of such phenotypically resistant bacteria, including persisters, stationary-phase bacteria, as well as those in the depths of biofilms. However, each of these types is characterised by its low rate of growth compared to log-phase bacteria under the same conditions. Nutritional starvation and high cell densities are also common characteristics of such bacteria.
Although resistant to antimicrobial agents in their slow-growing state, phenotypically resistant bacteria differ from those that are genotypically resistant in that they regain their susceptibility to antimicrobials when they return to a fast-growing state (e.g. when nutrients become more readily available to them).
The presence of phenotypically resistant bacteria in an infection leads to the need for prolonged courses of antimicrobial agents, comprising multiple doses. This is because the resistant, slowly multiplying bacteria provide a pool of “latent” organisms that can convert to a fast-growing state when the conditions allow (thereby effectively re-initiating the infection). Multiple doses over time deal with this issue by gradually killing off the “latent” bacteria that convert to “active” form.
However, dealing with “latent” bacteria by administering prolonged courses of antimicrobials poses its own problems. That is, prolonged exposure of bacteria to suboptimal concentrations of antimicrobial agent can lead to the emergence of genotypically resistant bacteria, which can then multiply rapidly in the presence of even high concentrations of the antimicrobial.
Long courses of antimicrobials are more likely to encourage the emergence of genotypic resistance than shorter courses on the grounds that non-multiplying bacterial will tend to survive and, interestingly, probably have an enhanced ability to mutate to resistance (Proc. Natl. Acad. Sci. USA 92, 11736-11740 (1995); J. Bacteriol. 179, 6688-6691 (1997); and Antimicrob. Agents Chemother. 44, 1771-1777 (2000)).
In the light of the above, a new approach to combating the problem of bacterial resistance might be to select and develop antimicrobial agents on the basis of their ability to kill “latent” microorganisms. The production of such agents would allow, amongst other things, for the shortening of chemotherapy regimes in the treatment of microbial infections, thus reducing the frequency with which genotypical resistance arises in microorganisms.
International Patent Application, Publication Number WO2000028074 describes a method of screening compounds to determine their ability to kill clinically latent microorganisms. Using this method, the Applicant has observed that many conventional antimicrobial agents, such as co-amoxiclav, azithromycin, levofloxacin, linezolid and mupirocin, which otherwise exhibit excellent biological activity against log phase (i.e. multiplying) bacteria, exhibit little or no activity against clinically latent microorganisms. This observation has necessitated the development of novel antimicrobials which may be used to kill clinically latent microorganisms.
International Patent Application, Publication Numbers WO2007054693, WO2008117079 and WO2008142384 describe compounds which exhibit biological activity against clinically latent microorganisms. Examples of such compounds include 4-methyl-8-phenoxy-1-(2-phenylethyl)-2,3-dihydro-1H-pyrrolo[3,2-c]quinoline, 4-(3-benzylpyrrolidin-1-yl)-2-methyl-6-phenoxyquinoline, N-[4-(3-benzylpyrrolidin-1-yl)-2-methylquinolin-6-yl]benzamide and pharmaceutically acceptable derivatives thereof.
Co-amoxiclav is a fixed dose combination product which consists of amoxicillin (in trihydrate form) and clavulanic acid (in the form of potassium clavulanate).